Presentations on site characterization and on other tools (SpatialDev).
Site characterization
Stratification of programme sites based initially on farming system, rainfall, elevation (proxy for temperature), population density and access to markets. In some areas, slopes are also important (e.g. in Ethiopian Highlands).
Q: What was the reason behind the concentration on wheat areas in Ethiopia?
A: I heard it was an emphasis from USAid and the national Gov't of Ethiopia might have insisted on other crops e.g. teff.
Q: Malawi was missing.
A: The concept of Africa RISING as a project is that it should be started in program areas and the result of this work would feed out similar agro-ecological zones. The idea was to focus on these megasites. Malawi is in the extrapolation zone. But it's unclear to me how it fits with the extrapolation zone - which is something that should be planned in the subsequent years.
The bulk of the work will be done in Tanzania and some work in Malawi. Mozambique was also one of the initial countries planned and it dropped out.
Comment: It would be good to characterize the new sites in Malawi now. These sites are in the right ecozone.
We are also working with CIMMYT in KotaKota in Malawi.
Q: Do you use soil as one characteristic? Soil brings about different crops.
A: We could have included soils but existing soil databases are not sufficiently detailed. We could do that.
--> On soils, at district level, you don't want to use soil as a factor. When you go down to village level, a simple DM analysis is enough to understand the distribution.
--> For uniformity, we would have to use the FAO soil database which is very general.
--> There are some national databases that are excellent e.g. in Ghana but they're not harmonized. That harmonization issue and the issue of scale are a problem to use this.
--> We can use soils only once we start working on-farm. Soils are mostly about management...
--> We still need to do some ground-truthing. I'm skeptical as re: quality of African databases. We need to do our own ground-truthing.
Population maps have been revised based on census data and refined on the basis of satellite imagery (i.e. independent checks). There are now good algorithms based on relief maps to estimate rainfall.
Q: Which data do we have?
A: Satellite data layers and interpolation from data stations e.g. CIU and another German global database. There are more stations that could be pulled in this. Some have a longer period (e.g. 100-200 years) and others are more recent or up-to-date but with lower resolution etc.
Soil data.
SpatialDev monitoring tools
PMT (working with login, request account and email feedback). This spatial mapping application can be coupled to infographics reports.
Data sources: where to get information on megasites, sites, strata, quick wins, surveys etc. and how to ingest data?
Q: Who has access to it? Who is the gatekeeper? (you need to interact with us to understand the context etc.).
A: It hasn't been discussed in this project but in other projects, the information is available for the organization as they like and it can be downloaded easily.
Comment: This presentation relates to Africa RISING, not Feed the Future (there is a FtF global database)...
Q: Is it technically possible to have different levels of access?
A: Sure, we can build more types, have private access...
Q: In the UK you wouldn't be able to share details about people online, you would have a random number.
A: We don't want to publish information about people
Comment: When we share data, we remove the name of the household etc. so we can ethically deal with it.
Q: Having only numbers is not enough.
A: We would indeed some description of the figures etc. There are placeholders for description etc.
Comment: This tool was originally developed for quick wins but we thought it could also be useful for the future. It's important you can check how you can input data. Farmers won't be able to access it but this was meant for us to access it.
Q: I would like to see the conceptual framework that guides this tools - how was it conceptualized to add/refine components.
A: It should indeed be seen as a software development project where we collect feedback and ideas from everyone.
--> This tool is meant to help with stratification etc. It could help us with raster layers displayed etc.
Q: Would it be possible to interact with layers? If we want to see the differences of rainfall etc. will there be processing capabilities.
A: Yes, we should be able to click on any layer and we can work on processing capabilities but it's not yet implemented.
Q: We need the data model and the data structure - what relates to what, what is disaggregated etc.
A:
Q: USAid and FtF have set up their own platform and they're conversant - they can easily share project data. It would be great to be able to share data with those other systems.
A: USAid wants to export the data to their systems etc. There are web services that can do it but that's another component.
--> We can integrate this with the FtF indicators too.
--> You will have to set up these indicators anyhow in a way or another. And we should focus on single data entry.
Q: Any system like this becomes frustrating if the data is not up-to-date. Who would be responsible for entering data and for making sure the data has been entered and what kind of pressure would be required to ensure it's entered?
A: The M&E focal points should ensure that this happens.
--> It should be clear what kind of information needs to be added etc. If we have that kind of plan it'll be much easier to update the system and then we can take stock of that data on a quarterly basis.
Q: In the USAid missions, there is a US-based firm to take care of data management. Nafaka is also running surveys in villages. Do we know what they're planning to do?
A: Not yet. But the collaboration with Nafaka is not yet in place... USAid should give feedback about the organization to run evaluations of FtF programs. IFPRI will be conducting surveys for AfricaRISING but not with the help of feedback.
Africa RISING M&E Expert Meeting
5-7 September 2012Large auditorium,ILRI Ethiopia, Addis Ababa
Back to the event agenda
Data platform session
Presentations on site characterization and on other tools (SpatialDev).Site characterization
Stratification of programme sites based initially on farming system, rainfall, elevation (proxy for temperature), population density and access to markets. In some areas, slopes are also important (e.g. in Ethiopian Highlands).Q: What was the reason behind the concentration on wheat areas in Ethiopia?
A: I heard it was an emphasis from USAid and the national Gov't of Ethiopia might have insisted on other crops e.g. teff.
Q: Malawi was missing.
A: The concept of Africa RISING as a project is that it should be started in program areas and the result of this work would feed out similar agro-ecological zones. The idea was to focus on these megasites. Malawi is in the extrapolation zone. But it's unclear to me how it fits with the extrapolation zone - which is something that should be planned in the subsequent years.
The bulk of the work will be done in Tanzania and some work in Malawi. Mozambique was also one of the initial countries planned and it dropped out.
Comment: It would be good to characterize the new sites in Malawi now. These sites are in the right ecozone.
We are also working with CIMMYT in KotaKota in Malawi.
Q: Do you use soil as one characteristic? Soil brings about different crops.
A: We could have included soils but existing soil databases are not sufficiently detailed. We could do that.
--> On soils, at district level, you don't want to use soil as a factor. When you go down to village level, a simple DM analysis is enough to understand the distribution.
--> For uniformity, we would have to use the FAO soil database which is very general.
--> There are some national databases that are excellent e.g. in Ghana but they're not harmonized. That harmonization issue and the issue of scale are a problem to use this.
--> We can use soils only once we start working on-farm. Soils are mostly about management...
--> We still need to do some ground-truthing. I'm skeptical as re: quality of African databases. We need to do our own ground-truthing.
Population maps have been revised based on census data and refined on the basis of satellite imagery (i.e. independent checks). There are now good algorithms based on relief maps to estimate rainfall.
Q: Which data do we have?
A: Satellite data layers and interpolation from data stations e.g. CIU and another German global database. There are more stations that could be pulled in this. Some have a longer period (e.g. 100-200 years) and others are more recent or up-to-date but with lower resolution etc.
Soil data.
SpatialDev monitoring tools
PMT (working with login, request account and email feedback). This spatial mapping application can be coupled to infographics reports.Data sources: where to get information on megasites, sites, strata, quick wins, surveys etc. and how to ingest data?
Q: Who has access to it? Who is the gatekeeper? (you need to interact with us to understand the context etc.).
A: It hasn't been discussed in this project but in other projects, the information is available for the organization as they like and it can be downloaded easily.
Comment: This presentation relates to Africa RISING, not Feed the Future (there is a FtF global database)...
Q: Is it technically possible to have different levels of access?
A: Sure, we can build more types, have private access...
Q: In the UK you wouldn't be able to share details about people online, you would have a random number.
A: We don't want to publish information about people
Comment: When we share data, we remove the name of the household etc. so we can ethically deal with it.
Q: Having only numbers is not enough.
A: We would indeed some description of the figures etc. There are placeholders for description etc.
Comment: This tool was originally developed for quick wins but we thought it could also be useful for the future. It's important you can check how you can input data. Farmers won't be able to access it but this was meant for us to access it.
Q: I would like to see the conceptual framework that guides this tools - how was it conceptualized to add/refine components.
A: It should indeed be seen as a software development project where we collect feedback and ideas from everyone.
--> This tool is meant to help with stratification etc. It could help us with raster layers displayed etc.
Q: Would it be possible to interact with layers? If we want to see the differences of rainfall etc. will there be processing capabilities.
A: Yes, we should be able to click on any layer and we can work on processing capabilities but it's not yet implemented.
Q: We need the data model and the data structure - what relates to what, what is disaggregated etc.
A:
Q: USAid and FtF have set up their own platform and they're conversant - they can easily share project data. It would be great to be able to share data with those other systems.
A: USAid wants to export the data to their systems etc. There are web services that can do it but that's another component.
--> We can integrate this with the FtF indicators too.
--> You will have to set up these indicators anyhow in a way or another. And we should focus on single data entry.
Q: Any system like this becomes frustrating if the data is not up-to-date. Who would be responsible for entering data and for making sure the data has been entered and what kind of pressure would be required to ensure it's entered?
A: The M&E focal points should ensure that this happens.
--> It should be clear what kind of information needs to be added etc. If we have that kind of plan it'll be much easier to update the system and then we can take stock of that data on a quarterly basis.
Q: In the USAid missions, there is a US-based firm to take care of data management. Nafaka is also running surveys in villages. Do we know what they're planning to do?
A: Not yet. But the collaboration with Nafaka is not yet in place... USAid should give feedback about the organization to run evaluations of FtF programs. IFPRI will be conducting surveys for AfricaRISING but not with the help of feedback.